A complete understanding of the function of RNA molecules requires a knowledge of the higher order structures (2D and 3D) as well as the characteristics of their primary sequence. RNA structure is important for many functions, including regulation of transcription and translation, catalysis, and transport of proteins across membranes. The understanding of these functions are important for basic biology as well as for bioassays and the development of drugs that can intervene in cases where pathological functionality of these molecules occurs. This may include viruses such as HIV or expression pathways in malignant cells.We have developed and continue to improve upon an RNA folding technique that uses concepts from genetic algorithms running on massively parallel SIMD (16,384 processors) and MIMD supercomputers. The algorithm has been recently modified so that it can run on a single processor of and SGI OCTANE, a 64 processor SGI/CRAY ORIGIN 2000 as well as a 512 processor CRAY T3E. The algorithm scales extremely well and is capable of running with hundreds of thousands of virtual processors. We are able to predict RNA pseudoknots. In addition, it contains other features such as a Boltzmann relaxation technique, a choice of different energy rules, forced and inhibited embedding of helical stems and the visualization of folding dynamics in real time.STRUCTURELAB, the heterogeneous bioinformatical RNA analysis workbench, which permits the use of a broad array of approaches for RNA structure analysis, has been enhanced in conjunction with RNA 2D3D and other computational tools such as AMBER to produce refined 3D atomic coordinates of RNA structures along with the visualization of these structures. Also, a novel interactive visualization methodology, STEM TRACE has been enhanced in STRUCTURELAB. This methodology enables the comparison and analysis of multiple sequence RNA folds from a phylogenetic point of view, thus allowing improvement of predicted structural results across a family of sequences. In addition, it permits the visualization of folding pathways when used in conjunction with the genetic algorithm. It has also been integrated with a motif finding capability, so that families of RNA sequences can be explored for common structural motifs. These systems have been adapted to other environments inside and outside our laboratory and NIH and are available upon request. Work has also begun on a WEB interface to these methodologies.This system is currently being employed in studying the HIV dimerization region, the genetic mechanisms of the viral cardiovirulence phenotype of the coxsackie B viruses which includes mechanisms for transcription attenuation, and the RNA/protein structural interactions between the enzyme thymidylate synthase (TS) and TS mRNA. Experiments have indicated that translation of human TS mRNA is controlled by an a negative autoregulatory mechanism with its own synthesized protein. The existence of these RNA/protein complexes have been found in human colon cancer cells.Z01 BC 08382-16 - databases, HIV, molecular dynamics, molecular structure, Parallel Computing, RNA folding, Supercomputing, Thymidylate Synthase, Coxsackie B Viruses, Computer Graphics, Computational methods, computational biology, Computer analysis, computer modeling, Computer Program,